© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 6485
    Prev
    Next
    INDEX
    Explanations

    Introductions to text

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    äºĨä¸Ģ份
    -0.25
    å±Ĩ
    -0.24
     backwards
    -0.24
    èİĬ
    -0.24
    _struct
    -0.24
    _STRUCT
    -0.23
    campo
    -0.23
    æ¯ĶæĪij
    -0.23
    translate
    -0.23
     Struct
    -0.22
    POSITIVE LOGITS
    ä¹ĭä¸Ģ
    0.30
    successfully
    0.29
    çļĦæĪIJåĬŁ
    0.26
    cth
    0.25
    èĥĮæĻ¯
    0.25
    ä½ľåĵģåĨħ容
    0.24
     Tee
    0.24
    çļĦç»ıåħ¸
    0.24
    æĹģ
    0.24
     used
    0.23
    Activations Density 0.007%

    No Known Activations